Complementary effects of adaptation and gain control on sound encoding in primary auditory cortex

Abstract

An important step toward understanding how the brain represents complex natural sounds is to develop accurate models of auditory coding by single neurons. A common model for auditory coding is the linear-nonlinear spectro-temporal receptive field (LN model). The LN model accounts for many features of auditory tuning, but it cannot account for long-lasting effects of sensory context on sound-evoked activity. Two mechanisms that may support these contextual effects are short-term plasticity (STP) and contrast-dependent gain control (GC), each of which has inspired an expanded version of the LN model. Both of these models improve performance over the LN model, but they have never been compared directly. Thus, it is unclear whether they account for distinct processes or describe the same phenomenon in different ways. To address this question, we recorded activity of neurons in primary auditory cortex of awake ferrets during presentation of natural sounds. We then fit models incorporating one nonlinear mechanism (GC or STP) or both (GC+STP) using this single dataset, and measured the correlation between the models predictions and the recorded neural activity. Both the STP and GC models performed significantly better than the LN model, but the GC+STP model performed better than either individual model. We also quantified the similarity between STP and GC model predictions and found only modest equivalence between them. Similar results were observed for a smaller dataset collected in clean and noisy acoustic contexts. These results suggest that the STP and GC models describe distinct, complementary processes in the auditory system. Significance StatementComputational models are used widely to study neural sensory coding. However, models developed in separate studies are often difficult to compare because of differences in stimuli and experimental preparation. This study develops an approach for making systematic comparisons between models that measures the net benefit of incorporating additional nonlinear elements into models of auditory encoding. This approach was then used to compare two different hypotheses for how sensory context, that is, slow changes in the statistics of the acoustic environment, influences activity in auditory cortex. Both models accounted for complementary aspects of the neural response, indicating that a hybrid model incorporating elements of both models provides the most complete characterization of auditory processing.